摘要
目的:在计算机辅助诊断中,对已提取分割好的CT肝脏组织中的兴趣区的几种提取方法做比较。方法:利用Otsu分割算法、梯度分割算法、指数熵分割算法、均值聚类分割算法和二维直方图分割算法,对已分割的CT肝脏组织的兴趣区进行自动分割,并和医生手工分割的图像相对比。结果:通过实验及对实验数据进行分析,找出了具有良好获取兴趣区能力的分割算法。结论:通过手工分割图像分别与Otsu分割算法、梯度分割算法、指数熵分割算法、均值聚类分割算法和二维直方图分割算法做对比,总结几种分割方法的有效性,为计算机辅助诊断起到了重要的作用。
Objective To compare several CT segmentation algorithms for extraction of liver region of interest. Methods The segmentation results by Otsu algorithm, gradient algorithm, index entropy algorithm, means clustering algorithm and 2D histogram algorithm were compared with that by manual segmentation, Results The segmentation algorithm with the best ability to obtain region of interest was found. Conclusion Manual segmentation is compared with several algorithms, including Otsu algorithm, gradient algorithm, index entropy algorithm, means clustering algorithm and 2D histogram algorithm, to sum up the effectiveness of above algorithms.
出处
《医疗卫生装备》
CAS
2012年第1期22-25,共4页
Chinese Medical Equipment Journal
基金
咸宁学院校级科研项目(KY10054)
关键词
OTSU分割
梯度分割
指数熵分割
均值聚类分割
二维直方图分割
兴趣区提取
CT图像
Otsu segmentation
gradient segmentation
index entropy segmentation
means clustering segmentation
2D histogram segmentation
region of interest extraction
CT image
